Immune tolerance can be broken and tumor rejection elicited in experimental animals by vaccines that induce antibodies against defined carbohydrate and peptide antigens and by vaccines that activate T-lymphocytes. The target antigens and optimal design for these two types of vaccines are generally quite different. Immunotherapy of ovarian cancer should ideally utilize both types of vaccines, however; we will initially focus on each separately. This Project is in transition from an exclusive focus on vaccines that induce antibodies to a more broad approach. In the current application we perform a final Phase II trial with the resulting antibody-inducing polyvalent vaccine and shift our focus to T-cell-inducing vaccines. Vaccines that induce antibodies are more advanced, largely because of the long availability of serologic assays to monitor vaccine development. Consequently, during the current interim funding period we will conclude a series of pilot trials with monovalent conjugate vaccines in preparation for a Phase II trial with our polyvalent KLH-conjugate vaccine. This trial is based on our unique expertise and experience in carbohydrate synthesis, chemical conjugation of carbohydrates and peptides to carrier proteins, and immunological adjuvants. In Aim 1, we will perform the serological analysis on the Phase II trial conducted under the fourth Project. The primary objective of the trial is to determine the clinical consequences of inducing antibodies against six to eight ovarian cancer antigens (GM2, globo H, Le-y, sTn, TF, MUC1, and possibly MUC 16 and KSA) widely expressed on the ovarian cancer cell surface. The secondary objectives of the trial are to a) confirm the potency of different batches of vaccine prepared over time, and to determine the correlation between antibody response against particular antigens and b) clinical course or c) antigens expressed in ovarian cancer recurrences. These three secondary objectives will be addressed in this Project. Recently, assays capable of detecting vaccine-induced T-cell reactivity without the need for prolonged in vitro sensitization have been developed. This now permits us to follow the same step-by-step process used in developing antibody-inducing vaccines. We will build on our unique experience with these assays, our experience with the CA125/MUC16 and Wilm's Tumor Antigen (WT1) antigens, and with the use of heteroclitic peptides, electroporation, and a variety of approaches designed to augment the immunogenicity of DNA vaccines. We will initiate this process with a series of pilot trials comparing different approaches to augmenting the immunogenicity of CA125/MUC16 and WT1. Selection of the approaches to be tested in these trials will be based in part on results of ongoing preclinical studies and clinical trials. In Aim 2 we will focus on heteroclitic peptides, human MUC16 B4 fraction protein and murine WT1 protein. In Aim 3 we will test DNA vaccines coding for human, xenogeneic or heteroclitic proteins, C3d and DHBc fusion proteins, and administration with electroporation or DNA coding for GM-CSF.